Analysis of Student Procrastinatory Behavior in Virtual Learning Environments Using Machine Learning

Abstract

Virtual Learning Environments (VLEs) are increasingly being used due to the uncertain global situation caused by COVID-19. Some major impacts of the pandemic are lockdowns and social distancing. This has led to academia taking measures to normalize the increased use of VLEs for teaching. However, this shift from face-to-face to online environments comes with many concerns—one of which is student procrastination. Procrastination is an observed student behavior of delaying tasks, which results in poor learning and can adversely affect student performance. Hence, is it essential to analyze student procrastination and determine the factors that cause this behavior whilst learning in VLEs. In this study, we present our findings on the impact of VLEs on student procrastination. We also analyze the performance of machine learning approaches to avoid manual intervention. We initially performed data collection, producing a dataset which was annotated by an expert, allowing us to visualize the pattern of student procrastination. Results confirmed the application of machine learning techniques for analysis for student’s behavior. The results demonstrated the effectiveness of supervised approaches with an accuracy of up to 83%. In contrast, unsupervised approaches do not seem appropriate for this task. We hope that future work based on this study will allow automatic data annotation based on a trained machine learning model. The findings of our work will help identify students prone to procrastinating and allow intervention to maintain their academic performance. The expected implication of the presented study is an improvement of educational practices, helping teachers and demonstrators to gain a better understanding of students’ behaviors.

Keywords: virtual learning environment, distance learning, machine learning, artificial neural networks, multi-layer perceptron, procrastination.